StataPresspresents
Discovering Structural Equation Modeling Using Stata
For more details or to order, visit us online atstata-press.com/discovering-sem.
By Alan C. Acock
Publisher: Stata Press Copyright: 2013 ISBN-13: 978-1-59718-133-4 Pages: 304; paperback Price: $48.00
How to contact us
Stata Press4905 Lakeway DriveCollege Station, TX 77845-4512USA
800-782-8272 (USA) 800-248-8272 (Canada) 979-696-4600 (Worldwide)
© Copyright 2013 StataCorp LP
1 Introductiontoconfirmatoryfactoranalysis
2 Usingstructuralequationmodelingforpathmodels
3 Structural equation modeling
4 Latent growth curves
5 Group comparisons
6 Epilogue—what now?
A Thegraphicaluserinterface
B Enteringdatafromsummarystatistics
Brief contents
Discovering Structural Equation Modeling Using Stata, by AlanC.Acock,successfullyintroducesboththestatisticalprinciples involved in structural equation modeling (SEM) andtheuseofStatatofitthesemodels.Thebookusesan application-based approach to teaching SEM. Acock demonstrateshowtofitawidevarietyofmodelsthatfallwithintheSEMframeworkandprovidesdatasetsthatenablethereadertofollowalongwitheachexample.Aseachtypeofmodelisdiscussed,conceptssuchasidentification,handlingofmissingdata,modelevaluation,and interpretation are covered in detail.
InStata,structuralequationmodelscanbefitusingthecommandlanguageorthegraphicaluserinterface(GUI)forSEM,knownastheSEMBuilder.Thebookdemonstratesbothoftheseapproaches.Throughoutthetext,theexamplesusethesem command. Each chapter alsoincludesbriefdiscussionsondrawingtheappropriatepathdiagramandperformingestimationfromwithintheSEMBuilder.Amorein-depthcoverageoftheSEMBuilderisgiveninoneofthebook’sappendixes.
ThefirsttwochaptersintroducethebuildingblocksofSEM.Chapter1beginswithoverviewsofCronbach’salphaasameasureofreliabilityandofexploratoryfactoranalysis.Then, building on these concepts, Acock demonstrates how toperformconfirmatoryfactoranalysis,discussesavarietyofstatisticsavailableforassessingthefitofthemodel,andshowsamoregeneralmeasurementofreliabilitythatisbasedonconfirmatoryfactoranalysis.Chapter2focusesonusingSEMtoperformpathanalysis.Itincludesexamplesofmediation,moderation,cross-laggedpanelmodels,andnonrecursive models.
Chapter 3 demonstrates how to combine the topics covered inthefirsttwochapterstofitfullstructuralequationmodels.Theuseofmodificationindicestoguidemodelmodificationandcomputationofdirect,indirect,andtotaleffectsforfullstructuralequationmodelsarealsocovered.
Chapter4detailstheapplicationofSEMtogrowthcurvemodeling.Afterintroducingthebasiclinearlatentgrowthcurvemodel,Acockextendsthistomorecomplexcasessuchastheinclusionofquadraticterms,time-varyingcovariates, and time-invariant covariates.
Inchapter5,Acockdiscussestestingfordifferencesacrossgroups in SEM. He introduces the specialized semsyntaxformultiple-groupmodelsanddiscussestheintricaciesoftestingforgroupdifferencesforthedifferenttypesofmodels presented in the preceding chapters.
Discovering Structural Equation Modeling Using Stata is anexcellentresourcebothforthosewhoarenewtoSEMandforthosewhoarefamiliarwithSEMbutnewtofittingthesemodelsinStata.ItisusefulasatextforcoursescoveringSEMaswellasforresearchersperformingSEM.
Comment from the Stata technical groupAbout the author
Alan Acock is a sociologist and a University Distinguished ProfessorintheSchoolofSocialandBehavioralHealthSciences at Oregon State University. Alan was also recognizedastheAlumniDistinguishedProfessorbasedon his work with students. He has published more than 130 articles in leading journals across the social and behavioral sciences, including Structural Equation Modeling, Psychological Bulletin, Multivariate Behavioral Research, Journal of Gerontology, Journal of Adolescence, American Journal of Public Health, American Sociological Review, Journal of Marriage and Family, Social Forces, Educational and Psychological Measurement, Journal of Politics, Prevention Science, American Journal of Preventive Medicine, and many others. He also authored thetextA Gentle Introduction to Stata.
The book’s audience
This book will appeal to
• Researchers new to structural equation modeling
• ResearchersfamiliarwithSEMwhowanttolearnaboutStata’sSEMcapabilities
• StudentswantinganappliedtreatmentofSEM,focusedonperforminganalysiswithStatasoftware
• Statauserswantingadditionalinformationaboutcommandsforstructuralequationmodeling